It really is too bad about DoNotPay... Because having a publicly available chatbot to navigate all the hoops companies make us jump through (which of course they set up on purpose in order to keep us locked into products, paying higher prices etc etc) is actually a great idea and would be an excellent, welfare increasing use of AI. I cou…
It really is too bad about DoNotPay... Because having a publicly available chatbot to navigate all the hoops companies make us jump through (which of course they set up on purpose in order to keep us locked into products, paying higher prices etc etc) is actually a great idea and would be an excellent, welfare increasing use of AI. I could see a similar approach being used to help people apply for public benefits they are entitled to. This kind of thing could actually could help a ton of people.
And the thing is, helping someone navigate customer service or apply for benefits does not require a law license! You could make a great product with this same basic business model and never run into the legal issues.
But then the companies will set up their own chatbots to make customer's chatbots jump through hoops, and it will come down to who has the best chatbot.
I was listening to a tech guy talk about how great it is that doctors can use chatbots to write letters to insurance companies about why a patient needs a particular treatment and it's like, yeah that's great until the insurance company gets a better chatbot.
The statement of care applications are rarely adversarial. They're a requirement to justify care. They lie pretty squarely in the domain of tedious, wrote work where automation could definitely decrease toil for doctors.
Think of it like a letter from your doctor for your teacher. It's super rare they ever challenge it, the fact that it's from your doctor is the main requirement.
It really is too bad about DoNotPay... Because having a publicly available chatbot to navigate all the hoops companies make us jump through (which of course they set up on purpose in order to keep us locked into products, paying higher prices etc etc) is actually a great idea and would be an excellent, welfare increasing use of AI. I could see a similar approach being used to help people apply for public benefits they are entitled to. This kind of thing could actually could help a ton of people.
And the thing is, helping someone navigate customer service or apply for benefits does not require a law license! You could make a great product with this same basic business model and never run into the legal issues.
But then the companies will set up their own chatbots to make customer's chatbots jump through hoops, and it will come down to who has the best chatbot.
I was listening to a tech guy talk about how great it is that doctors can use chatbots to write letters to insurance companies about why a patient needs a particular treatment and it's like, yeah that's great until the insurance company gets a better chatbot.
The statement of care applications are rarely adversarial. They're a requirement to justify care. They lie pretty squarely in the domain of tedious, wrote work where automation could definitely decrease toil for doctors.
Think of it like a letter from your doctor for your teacher. It's super rare they ever challenge it, the fact that it's from your doctor is the main requirement.
It sounds like it was a really useful service at first.
💯